Research in functional programming on multiprocessor architectures. Final report
The importance of parallel computing hardly needs emphasis. It is clear that many physical problems and abstract models are seriously compute-bound, since sequential computer technology is now facing fundamental physical limitations that appear insurmountable. It is widely believed that the only feasible path toward higher performance is to consider radically different computer organizations, in particular ones exploiting parallelism. This argument has been made repetitively for many years now, and considerable progress has been made in the construction of highly-parallel machines. For example, one of the simplest and most promising types of parallel machines is the well-known multiprocessor architecture which can be characterized as a large collection of simple, autonomous processors with either shared or distributed memory, and usually {open_quotes}communicating by messages.{close_quotes} Although designing and building such machines has proceeded at a dramatic pace, the development of effective ways to program them generally has not. This is an unfortunate state of affairs, since experience on sequential machines tells one that software development is the most critical element is a system`s design, not the hardware. The immense complexity of parallel computation can only increase this dependence on software - clearly one needs effective ways to program these parallel machines. This research project was concerned with precisely this problem, and in this final report the author summarizes advances made in language design models for parallel computing, and semantic program analysis. His overall methodology - para-functional programming - is based on a functional programming model augmented with tools to allow the programmer to map programs to specific multiprocessors.
- Research Organization:
- Yale Univ., New Haven, CT (United States). Dept. of Computer Science
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- FG02-86ER25012
- OSTI ID:
- 10194234
- Report Number(s):
- DOE/ER/25012-6; ON: DE95002808; TRN: 94:010352
- Resource Relation:
- Other Information: PBD: Apr 1993
- Country of Publication:
- United States
- Language:
- English
Similar Records
Automatic synthesis of concurrent control for multiprocessor systems of general topology through fine-grain mapping
Parallel methods for the flight simulation model